How Sports Teams Use AI to Win — and What Owners Can Learn
In 2002, a cash-strapped baseball team in Oakland figured out that the statistics everyone else ignored could beat the statistics everyone else worshipped. The Oakland A's rode that insight to a record-tying win streak on one of the smallest payrolls in the league, "Moneyball" became a book and a movie, and every professional sport quietly took notes. Two decades later, that trickle has become a flood — and the people teams are hiring to ride it increasingly hold PhDs in physics, not front-office résumés.
The physicist who helped Liverpool win the league
The clearest example is Ian Graham. He holds a doctorate in physics from Cambridge, and in 2012 Liverpool FC hired him to build the first in-house analytics department in the Premier League. His team built models to value every action a player takes and to find undervalued talent the scouting consensus had missed — the data behind key signings and even a manager hire that would define an era. Over the next decade Liverpool won the Champions League and, in 2020, their first league title in 30 years. The press dubbed it "Merseyball." Graham later wrote a book about it, aptly titled How to Win the Premier League.
Every game is now a firehose of data
What makes today different from the original Moneyball era is the sheer volume of what teams can measure. Optical tracking systems record the position of every player and the ball dozens of times per second. In the NBA, that turns every pass, cut, and contest into data. In Major League Baseball, systems track the spin and movement of every pitch and the launch angle of every swing. Formula 1 cars stream thousands of telemetry channels in real time, and race strategy is now run by data scientists on the pit wall.
A human coach can't watch all of that. AI can. Machine-learning models sift the torrent for the patterns that matter: which lineup combinations quietly outscore opponents, which pitcher is tipping his pitches, which player's movement suggests an injury is coming before he feels it. The edge isn't the data — everyone has the data now. The edge is having something that can actually read it.
The edge isn't the data — everyone has the data now. The edge is having something that can actually read it.
It's not just the pros — colleges are all in
College programs have followed the same path, often with real budgets behind it. Athletic departments now employ performance analysts who work with tracking, video, and scouting data to shape tactics and player development, and use predictive models to identify recruits whose high-school performance is most likely to translate to the next level. Some of the cleverest work is off the field entirely: at Michigan State, an analytics staffer became known for building models that generate game schedules to maximize TV exposure while minimizing missed class time. The job listings tell the story — leagues and teams from MLB to MLS are actively hiring data scientists, and "SQL, Python, and AI" now sit on scouting-department job descriptions next to "knows the game."
What a business owner should take from this
It's easy to file this under "billionaires buying an edge." But the underlying move is available to any organization, and it's cheaper than it has ever been. Sports teams didn't win by collecting more data than their rivals — everyone in the league gets the same feeds. They won by being first to turn that data into decisions: who to sign, when to rest a player, which situation to attack.
Your business has its own version of tracking data. It's called your quote history, your invoices, your job notes, your customer records, your win-loss patterns. Most companies collect all of it and read almost none of it. The teams winning championships figured out that the insight was already sitting in the files — they just needed something tireless enough to watch every play. That's the same bet behind giving a company a brain that actually knows its own history rather than a chatbot that knows nothing about you. You don't need a Cambridge physicist. You need to stop leaving your own game film unwatched.